Exchange Rate Policies on the Last Stretch



Exchange Rate Policies

on the Last Stretch*

by

Jürgen von Hagen

ZEI, University of Bonn; Indiana University; and CEPR

and

Jizhong Zhou**

ZEI, University of Bonn

February 2003

1 Introduction

At the Copenhagen Summit on 12 and 13 December 2002, the European Council declared that accession negotiations had been successfully concluded with eight Central and Eastern European (CEE) accession countries—the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, the Slovak Republic, and Slovenia—as well as Cyprus and Malta. These countries will join the European Union (EU) from 1 May 2004.[1] Now that the accession process has reached its final phase, the conduct of monetary and exchange rate policies in the accession countries are confronted with new challenges. EU membership requires that accession countries abolish any remaining controls to allow free cross-boarder movement of capital, a requirement that has already largely been implemented. As a result, these countries will be fully exposed to the whims of international financial markets, and exchange rates are likely to be more volatile. However, according to Article 99 (1) of the Treaty Establishing the European Community, exchange rate policies are a matter of common concern once a country is admitted to the EU and the new members cannot treat their exchange rates with total neglect, even if they preferred to do so. This implies that exchange rate flexibility is constrained. As semi-fixed exchange rates tend to be less viable under free capital mobility,[2] the monetary authorities of the new EU members must find safe strategies to ensure exchange rate stability in an environment of high capital mobility.

Another challenge is related to the participation in Economic and Monetary Union (EMU). All accession countries aspire for an early EMU membership, and, therefore, must meet the convergence criteria as soon as possible. These criteria include price stability and exchange rate stability, which amounts effectively to the stipulation that real exchange rates must be stable, too. However, since the new member countries are catching up to the income level of the existing EU members, their price levels relative to the EU will also rise, so that a real appreciation of their currencies is inevitable, especially in the long run.[3] This is likely to create conflicts between price stability and exchange rate stability.

This paper addresses these challenges and analyses appropriate monetary and exchange rate policies for the eight CEE accession countries during the run-up to the EU and EMU. At the starting point, the existing exchange rate regimes vary across countries, ranging from rigid currency board arrangements to flexible floating regimes. Nevertheless, the end point is fixed for all these countries, namely the adoption of the common currency. Therefore, the transition strategies toward EMU are likely to differ across countries, despite the similarities of the catching-up process and the common same ultimate goal.

The paper is organized as follows. Section 2 provides a brief review of current exchange rate policies and convergence performance by the accession countries. In section 3, we analyze the implications for exchange rate policies of heightened capital mobility and trend real appreciations, two prominent issues that are likely to complicate the making of exchange rate policies. Section 4 discusses the standard route toward EMU via ERM-II regimes, while the non-standard approaches via currency boards or unilateral euroization are discussed in section 5. Section 6 concludes.

2 The Current Scenario: An Assessment

2.1 Current Exchange Rate Policies

Although all the eight CEE accession countries are aiming at the irrevocable pegging of their currencies to the Euro in the medium-term future, their current exchange rate regimes are quite diverse, reflecting differences in economic fundamentals, such as economic openness, size of economy, level of economic development, and trade structure, as well as prevailing macroeconomic circumstances when these regime choices are made, including reserve adequacy and external competitiveness.[4]

Table 1: Exchange Rate Regimes and Monetary Policy Frameworks (as of Nov. 2002)

|Coutrry |Exchange Rate Regime |Monetary Policy Framework |

|Czech Rep. |Independent float |Inflation targeta (Jan. 1998) |

|Estonia |Currency board (Euro) |Exchange rate target (Jun. 1992) |

|Hungary |Horizontal band ((15%, Euro) |Inflation target (May 2001) |

|Latvia |Conventional fixed peg (SDR) |Exchange rate target (Feb. 1994) |

|Lithuania |Currency board (Eurob) |Exchange rate target (Apr. 1994) |

|Poland |Independent float |Inflation target (Apr. 2000) |

|Slovak Rep. |Managed float |Multiple targets (Oct. 1998) |

|Slovenia |Managed float |M3 target (Oct. 1991) |

Source: IMF, IFS; updated based on information from national sources.

a Net inflation targets for 1998—2001; headline inflation targets since 2002.

b On February 2, 2002 the anchor currency was switched from the US dollar to the Euro.

The variety in the exchange rate regime choices also reflects different stabilization strategies and the availability of alternative monetary policy frameworks for this purpose. Achieving price stability still remains the main stabilization task.[5] For this purpose, the choice of a nominal anchor of monetary policy is important. Table 1 reports the exchange rate regimes and monetary policy frameworks currently adopted in the CEE accession countries. It shows that the countries use exchange rate targets, inflation targets, or monetary targets for that purpose.

The Exchange Rate as a Nominal Anchor

Only the three Baltic states — Estonia, Latvia, and Lithuania—currently maintain some type of exchange-rate targeting framework for their monetary policies. Estonia and Lithuania have currency board arrangements (CBA), the most rigidly fixed exchange rate regime short of monetary union. It should be noted that both countries are among the smallest countries in the CEE candidates.[6] This conforms well the international experiences that CBA tends to be adopted by small open economies.[7] Both CBAs now have the Euro as the anchor currency.[8] Latvia has a less rigid exchange rate anchor: a conventional fixed peg vis-à-vis the Special Drawing Right (SDR). Although the Latvian peg has been stable since its introduction in April 1994, the commitment to a stable exchange rate is not guarded by the institutional arrangement that we observe by a CBA, therefore, flexibility is still reserved to some extent. Moreover, pegging to a basket currency like the SDR can mitigate impacts on the home currency of exchange rate fluctuations among major international currencies. However, an immediate problem with this regime in the current accession scenario is that it is not tied to the Euro, which is regarded as incompatible with the spirit of integration into EMU.

Inflation Targeting with Flexible Exchange Rate Regimes

Inflation targeting has gained popularity in the CEE accession countries, especially in those more advanced in the transition process. The Czech Republic, Hungary, and Poland have abandoned exchange-rate targeting in favour of inflation targeting as the framework for monetary policies. While the Czech koruna and the Polish zloty float freely, the Hungarian forint fluctuates within an official band of (15%. Compared to exchange rate anchors, inflation targeting directly addresses the issue of price stability, which is explicitly stipulated as the primary objective of monetary policies. Stronger independence enables the central banks to devote the use of all available instruments to achieving the inflation targets, especially to counteract the excessive price rises due to expansionary fiscal policies.[9] Meanwhile, the exchange rates are flexible to absorb the impact of capital flows, which tends to be larger as the countries are further integrated into the international financial markets.

Implementing an inflation targeting strategy requires a functioning domestic financial sector to facilitate the transmission of interest rate policies, typically used to implement inflation targeting. Table 2 reports an index of financial development for the eight CEE accession countries. The index is based on the average values of the EBRD indices of banking and non-banking reforms, with higher index values denoting more advanced financial sector development. It is expressed as a percentage of the EU average.[10] The table shows that the Czech Republic, Hungary, and Poland, whose index values are close to the EU standard, have indeed more developed financial sectors than the other accession countries.

Table 2: Financial Development and Economic Openness

|Country |Index of Financial Developmenta |Economic Opennessb |

| |1999 |2001 |1999 |2001 |

|Czech Republic |73 |78 |50 |62 |

|Estonia |78 |78 |68 |70 |

|Hungary |85 |90 |55 |62 |

|Latvia |62 |65 |35 |37 |

|Lithuania |66 |70 |36 |45 |

|Poland |77 |81 |24 |25 |

|Slovak Republic |58 |65 |55 |67 |

|Slovenia |73 |70 |46 |52 |

Source: Own calculations based on EBRD, Transition Report; IMF, IFS.

a In percent of the EU average.

b The average of export and import values in percent of GDP.

Inflation targeting can be largely equivalent to exchange-rate targeting, if the economy is very open to foreign trade. To see this, we can decompose an inflation target ((TARGET) into a target for non-tradable goods inflation ((N), which is not influenced by foreign trade, and a target for tradable goods inflation ((T), which is influenced by the exogenous foreign inflation ((*) and a targeted rate of home currency depreciation ((e). That is,

(TARGET = (1 - ()(N + ((T = (1 - ()(N + (((* + (e),

with ( denoting the share of tradable goods in the consumer price index. The larger (, the more open an economy to foreign trade, the more important is the implicit exchange–rate target for inflation targeting. Interestingly, Table 2 indicates that two of the three current inflation targeters, Hungary and the Czech Republic, are more open to foreign trade than Latvia and Lithuania. This suggests that, even with an inflation targeting framework in place, both countries need to pay much attention to exchange rate stability.[11]

Managed Floating with a Monetary Target or Multiple Targets

Slovenia is the only country among the EU accession countries that currently uses a monetary target as the anchor for monetary policy. Its exchange rate regime is a managed float. In practice, Slovenia has used the monetary target pragmatically to take into account external competitiveness and sustainability.[12] While money growth targets are set to steer a disinflation process, the nominal exchange rate is heavily managed to avoid severe real exchange rate misalignment. As a result, Slovenia’s de-facto exchange rate regime has been among the most stable ones in the region. Similarly, the Slovak Republic pursues a managed floating exchange rate regime which allows sufficient room for policy manoeuvre. While monetary policy has been predominantly geared toward a core inflation goal, the monetary authority also considers exchange rate targets, and intervenes in the foreign exchange market to restrain excessive exchange rate volatility.

In sum, although official descriptions of monetary policy regimes among the accession countries show some divergence, exchange rate considerations are likely to play a large role for all of them except, perhaps, Poland. This suggests that the exigencies implied by considering their exchange rate policies as a matter of common concern will not place overwhelming demands on them.

2.2 Convergence Performance

Roughly ten years into transition, the CEE accession countries have made considerable progress in the convergence toward the EU standards. The upcoming EU accession is a recognition of these achievements and signifies that the accession criteria have been fulfilled.[13] However, since the accession countries aspire for EMU membership, they need to further meet the Maastricht Criteria. In this sub-section we will briefly assess the latest development in the real and nominal convergence of the eight CEE accession countries.

Table 3: Real Convergence

| |GDP per capita* |Productivity growth |

| |(% of EU average) |(% p.a.) |

|Country | | |

| |1997 |2001 |2002a |1996—2000b |2001 |2002a |

|Czech Republic |63 |57 |60 |2.0 |3.5 |4.2 |

|Estonia |37 |42 |38 |5.8 |5.4 |4.5 |

|Hungary |47 |51 |50 |3.1 |3.5 |4.3 |

|Latvia |27 |33 |32 |4.8 |7.6 |6.0 |

|Lithuania |31 |38 |31 |4.7 |6.3 |5.9 |

|Poland |40 |40 |40 |3.8c |3.2 |4.5 |

|Slovak Republic |47 |48 |50 |6.3 |2.2 |3.3 |

|Slovenia |68 |69 |73 |7.1 |2.5 |2.5 |

|EU-15 |100 |100 |100 |2.2 |0.4 |1.1 |

Sources: European Commission, Strategy Paper (2002) and Regular Report (1998); Deutsche Bank Research (2002); IMF, IFS; Buiter and Grafe (2001).

* In Purchasing Power Standards.

a Predictions made by Deutsche Bank Research (2002).

b Average productivity growth rates over the period 1996—2000.

c Average over 1996—1999.

Real Convergence

A common indicator of real convergence is the candidates’ per capita GDP (measured in purchasing power standards) relative to the EU average, which reflects the alignment of economic welfare in the accession countries to the level prevailing in the EU.[14] Table 3 shows that the degree of convergence is rather diverse in the accession countries. At the lowest end of the scale, per capita GDP of Latvia and Lithuania is less than one third of the EU average in 2002. At the highest position, Slovenia’s per capita GDP has risen to 73 percent of the EU average, above some poorest EU countries.[15] On average, this group of accession countries have a per capita GDP of roughly 47 percent of the EU level.[16]

From a dynamic point of view, most of the accession countries have seen a rise in their per capita GDP relative to the EU average over the period 1997—2001, except for the Czech Republic, which had a 6-percentage-point decline, and Poland, whose per capita GDP has been growing at exactly the same speed as that of a typical EU country. This catching-up process is mainly propelled by faster productivity growth in the accession countries than in the relatively richer EU countries. Table 3 shows that, over the period 1996—2000, only the Czech Republic had a productivity growth rate lower than that of the EU average, while the other countries were improving their labour productivity faster than the EU on average. In the last two years, all the eight accession countries exhibited faster productivity growth than a typical EU country.

Nominal Convergence

The “Maastricht Criteria” set the conditions for the nominal convergence for the accession into EMU. Although the accession countries will not join the Euro earlier than 2006, it is worthwhile checking to what extent the Maastricht Criteria have already been met as well as in which areas further efforts are necessary. Table 4 provides an overview.

The Maastricht Treaty stipulates that inflation rates should not exceed the average inflation rate of the three best performing EMU member states by more than 1.5 percentage points. This leads to a reference inflation rate of 3.4% per annum in 2001 and 3.0% in 2002. It is clear from Table 4 that only Latvia and Lithuania fulfilled this criterion in 2001. The latest data indicate that the Czech Republic and Poland also met the inflation criterion in 2002. For the other countries, especially Slovenia and Hungary, serious efforts still need to be taken to achieve low inflation rates.

On the fiscal front, the Maastricht Criteria set ceilings for general government budget deficits at 3% of GDP and for public debt stocks at 60% of GDP.[17] In 2001, four countries reported fiscal deficits larger than 3% of GDP, and three of them (the Czech Republic, Hungary, and Poland) failed to bring their deficits below the threshold in 2002. Hungary and Poland have the largest debt stocks in this group. None of the eight CEE accession countries, however, exceeds the 60-percent ceiling on public debt. This suggests that most accession countries have disciplined their fiscal policies in recent years, and fulfil or are close to fulfil the related convergence criteria in 2002. It should nevertheless be noted that the approaching EU accession may actually weaken or worsen their fiscal positions. Although the phasing-out of subsidies, tax harmonization, and transfers from the Cohesion Fund will have positive effects on the fiscal budget, they are insufficient to compensate the negative effects caused by contributions to the EU budget, preparation of matching funds for the projects supported by the EU funds, and other infrastructure expenditures (Kopits and Székely, 2002). It is estimated that the overall direct impact of EU accession on the fiscal budget of these countries is to increase deficits (or reduce surplus) by 3%—4.75% of GDP. This may deteriorate the fiscal balance of the accession countries to such an extent that almost all of them will fail on the fiscal test. Therefore, the countries need further fiscal consolidations.

Table 4: Nominal Convergence

| |Inflation |Fiscal |Public |Exchange |Interest |

| |rate |balance |debt |rate |rate |

| |(% p.a.) |(% of GDP) |(% of GDP) |fluctuations* |(% p.a.) |

|Country | | | | | |

| |2001 |2002 |2001 |

| |1998 |2000 |1998 |2001 |1998 |2001 |

|Czech Republic |5 |3 |6.6 |8.7 |12.0 |13.3 |

|Estonia |1 |3 |11.0 |9.7 |16.9 |24.0 |

|Hungary |9 |9 |4.3 |4.7 |12.1 |16.0 |

|Latvia |9 |2 |5.8 |2.3 |13.0 |16.4 |

|Lithuania |2 |2 |8.6 |3.7 |15.0 |8.2 |

|Poland |10 |10 |4.0 |3.2 |9.1 |11.4f |

|Slovak Republic |8 |7 |2.6 |6.3 |11.3 |24.8f |

|Slovenia |7 |6 |1.3 |1.9 |6.3 |9.2 |

|CEE averaged |6.4 |5.3 |5.5 |5.1 |12.0 |15.4 |

|EU averagee |2.1 |1.7 |-1.5 |-1.5 |20.6 |20.8 |

a Number of capital transaction categories subject to controls. The total number of categories is 11. Source: Own calculations based on IMF, AREAER (1998, 2000).

b Source: European Commission, Regular Report (2002).

c Gross capital flows is the sum of outflows and inflows of direct investment, portfolio investment, and other investment. Source: IMF, IFS.

d Simple average across 8 CEE accession countries.

e Number of capital controls is simple average across 15 EU member states; Net FDI flows and gross capital flows are GDP-weighted average for the Euro Area.

f Data for 2000.

From a medium to long term perspective, the CEE accession countries are expected to experience even more capital mobility in the future. On the de jure side, further capital account liberalization will allow more freedom in the cross-boarder movement of capital. On the de facto side, higher profitability associated with faster economic growth in the accession countries will attract more capital inflows, and the integration of these countries into the international financial market will lead to substantial increase in the gross capital flows.

Given high capital mobility, central banks are likely to be involved in large-scale foreign exchange interventions, if they wish to achieve exchange rate targets. Meanwhile, the probability of abandoning the exchange rate target is rising as the magnitude of capital flows surges. This is because a large, if not the largest part of all capital inflows take the form of portfolio investment or other financial investments, a substantial part of which is short-term (“hot money”). The experience of the 1990s shows that such capital flows can easily revert in the case of political or economic instability in the host country, swings in market expectations, or contagion of financial crises from abroad. When this happens, central bank interventions are limited by the availability of international reserves, which can be overwhelmed by the magnitude of capital outflows. In that case, a crash of an exchange rate regime of limited flexibility is inevitable. In case of capital inflows, sterilization is, in principle, not limited. However, sterilized intervention can be costly, and unsterilized intervention can lead to excessive inflation and, consequently real appreciation and a deterioration of external account. This is, in essence, the “convergence play” experienced by countries in the ERM of the late 1980s and early 1990s. Empirical studies show that the combination of large capital inflows and real exchange rate appreciations are typical leading indicators of currency crises in emerging economies.[20]

As a result, high capital mobility will make exchange rate stability an ever more difficult objective, unless monetary policies are fully devoted to this end. Large amounts of foreign capital expected to flow into the accession countries will push the national currencies toward appreciation. Exchange rate volatility tends to be increased, which can be avoided only by subordinating all monetary instruments toward the objective of exchange rate stability.

3.2 The Influence of Real Appreciations

The other major challenge to the monetary authorities in the CEE accession countries is to achieve exchange rate and price stability in the presence of real appreciations of their currencies. The real appreciation of the accession countries’ currencies vis-à-vis that of their main trading partner (i.e. the Euro) reflects the fact that the inflation differentials between the accession countries and the EU are larger than the depreciation of the candidates’ currencies against the Euro (if any) can offset. The common causes of higher inflation in the accession countries include the monetization of fiscal deficits and the adjustment of relative prices. Given the fact that the former cause becomes less likely as fiscal discipline is strengthened in the accession countries, the latter becomes increasingly prominent. The main driving force of relative price adjustment is productivity growth.[21] Since this is inevitable during the accession process, the monetary authorities face a trade-off between exchange rate stability and price stability.

The relationship between productivity growth and relative price changes is usually summarized in terms of the Balassa-Samuelson (B-S) effect, which refers to “a tendency for countries with higher productivity in tradables compared with non-tradables to have higher price levels” (Obstfeld and Rogoff, 1996, p. 210). The mechanism linking relative productivity growth with price level differentials is wage equalization across tradable and non-tradable sectors. The wage increase in the tradable sector caused by productivity growth pulls up the wage level of the non-tradable sector through wage equalization and inter-sectoral labour relocation. If productivity growth in the non-tradable sector is slower, the price of non-tradables will rise. Since the price of tradables is determined by the world price, assuming that the law of one price holds for tradables, the price of non-tradables relative to the tradables will rise, and so is it with the overall price level. If the relative productivity growth of the tradables as compared with non-tradables is faster at home than in abroad, home price level will rise faster than foreign, so a positive inflation differential will be observed.

The B-S effect is frequently cited as a factor underlying the inflation differentials of the CEE accession countries relative to the EU. Productivity growth in the tradable sector is faster than in the non-tradable sector in the accession countries, since modern production technology of the tradable sector can be easily transferred into the accession countries, leading to substantial improvement in labour productivity, while the non-tradable sector is, by definition, closed to foreign competitions.[22] Moreover, the productivity growth differential is more pronounced in the accession countries than in the EU due to the catching-up process. As a result, the accession countries’ currencies tend to appreciate in real terms vis-à-vis the Euro. If nominal exchange rates are fixed, these real appreciations lead to higher inflation in the accession countries than in the EU.

Table 6: Inflation Differentials due to the Balassa-Samuelson Effect in the 1990s

| |Consumer Price Inflation Differentialsa (in %| |

|Empirical Study |p.a.) |Sample |

|Fischer (2002) |2.4 — 3.1 |8 CEECs |

|Lommatzsch and Tober (2002) |1.1 — 4.1 |5 CEECs |

|De Broeck and Sløk (2001) |2.9 |10 CEECs |

|Coricelli and Jazbec (2001) |2.0 |10 CEECs, 9 other TEs |

|Halpern and Wyplosz (2001) |2.9 — 3.1b |8 CEECs, Russia |

|Rother (2000) |1.9 — 4.2 |Slovenia |

|Pelkmans et al. (2000) |3.8b |10 CEECs |

|Kovács and Simon (1998) |2.9b |Hungary |

Note: CEECs: Central and Eastern European Countries. TEs: Transition Economies.

a Own calculations for the case of fixed exchange rate regimes. Computed based on the average productivity differential of 7.3% per annum and the average share of tradable goods in consumer price index of 0.46. See the explanations in the text.

b Taken from original studies.

Empirical studies do find some support for the B-S effect in the CEE accession countries. Table 6 summarizes the main findings of a selection of studies by reporting the implied consumer price inflation differentials between the CEE accession countries and the EU if the former adopt fixed exchange rates against the Euro. For some studies we calculate the implied inflation differentials based on their coefficient estimates. These studies use, in essence, the following empirical approximation:

π - π* ( β(1 - α)(aT - aN),

where π (π*) denotes the home (foreign) consumer price inflation rate, α is the share of tradable sector in the consumer price index, and aT (aN) is the rate of productivity growth of the tradable (non-tradable) sector.[23] While some studies estimate β, some others estimate directly β(1 - α). For our calculation we gauge the annual relative productivity growth (aT - aN) at 7.3%[24] and the share of tradables in CPI (α) at 0.46.[25] The results reported in Table 6 show that the magnitude of inflation differentials that can be attributed to relative productivity growth differentials ranges from 1% to 4% in each year. An implication of these results is that accession countries maintaining a fixed Euro exchange rate may fail to meet the inflation criterion of the Maastricht Treaty, which allows, at maximum, a 1.5% inflation differential.[26] A careful balance must be sought between the two objectives of price and exchange rate stability.

There are several caveats to the above-mentioned analysis.[27] First of all, the B-S effect is a long-run tendency, which may be less prominent over a shorter time horizon. Second, productivity growth of the service sector, which is the main constituent of the non-tradable sector, can be very fast in the accession countries, as it is developing essentially from a very low starting level. Third, productivity growth of the tradable sector will gradually lose its momentum as it reaches higher level. Fourth, the pressure of high unemployment rates may prevent the wage rate from equalizing at a level compatible with tradable-sector productivity growth. All these factors tend to reduce the inflation differentials caused by the B-S effect. And after all, relative productivity growth in the EU may also offset to some extent the accession countries’ inflation differentials over the EU.[28]

Despite these qualifications, the B-S effect cannot be ignored as a source of inflation differentials. In the medium to long run, the productivity-driven relative price adjustment is the dominant factor of high inflation rates, which may constrain the exchange rate policies of the accession countries over a longer time span.

3.3 Policy Responses

Facing the challenges of high capital mobility, one response is to follow the “bi-polar” approach in the choice of exchange rate regimes.[29] At the one extreme end, more flexible exchange rate regimes, such as free or managed float, are adopted, which allow the exchange rate to adjust more freely to absorb the shocks caused by volatile capital flows. At the other extreme end, very rigidly fixed regimes, such as currency board arrangements, are adopted, which strengthen the commitment to exchange rate stability by devoting all monetary instruments to that objective. Another response is to conduct conservative fiscal policies, especially in the presence of capital inflows, which can prevent capital inflows from fuelling domestic price pressures and the build-up of external debts that would eventually threaten the financial system and the exchange rate arrangement.[30]

To cope with the challenges caused by real appreciations, one needs to find measures that reduce the magnitude of such real appreciations in the first place. One measure is to promote non-tradable sector development, especially the improvement of its productivity, so that those real appreciations attributable to the B-S effect is attenuated. Another measure is to strengthen fiscal discipline to contain aggregate demand, especially those on non-tradable goods, so that the relative price of non-tradable goods does not rise too fast.

These responses and measures, however, have their own shortcomings. Floating exchange rate regimes may result in volatile exchange rates, which have negative consequences for small open economies like the CEE accession countries. With hard pegs, the monetary authority loses control on domestic monetary supply and, henceforth, inflation. The measures dealing with real appreciations can only partially attenuate the magnitude of real appreciations, but are unable to solve the issue completely. The remaining real appreciation driven by relative productivity growth is an equilibrium phenomenon, which will be manifested by nominal appreciation, higher inflation rates, or both.

4. Entry into EMU via ERM-II

According to the official position of the EU, the procedure leading to the adoption of the Euro involves three steps: (1) entry into the EU; (2) participation in the ERM-II; and (3) entry into EMU after meeting the convergence criteria. While there is no separate national monetary and exchange rate policy after the entry into EMU, during the period leading to that third stage the monetary authorities of the CEE accession countries must still carefully formulate their exchange rate policies to facilitate a smooth accession to the EU and EMU. However, there are alternatives ways leading to EMU.

4.1 Exchange Rate Regimes Consistent with EMU Accession

Before joining the EU, the CEE accession countries have free choices as to their exchange rate regimes. Upon the accession, the new EU members will still have much freedom in choosing their exchange rate arrangements, except that they must now treat their exchange rate policies (as well as other economic policies) as a matter of common concern of all the EU members. This requirement is aimed at avoiding competitive devaluations among member countries. All exchange rate regimes satisfying this requirement are consistent with the EU accession, including those with fixed central parities (currency boards, pegged rates, horizontal bands) and those under which exchange rates are determined mainly by market forces without heavy manipulations of the authorities (managed or free float). From this perspective, the eight CEE accession countries that will join the EU in 2004 have now exchange rate regimes consistent with EU accession (see Table 1).

After joining the EU, all the new members are expected, sooner or later, to participate in the second stage of the Exchange Rate Mechanism (ERM-II) for at least two years before the final entry into EMU. The ERM-II is in essence a horizontal band regime where the Euro exchange rate of a member state currency is allowed to fluctuate within a (15% band around the fixed central parity. Several exchange rate regimes are excluded from the option list, including independent float (the Czech Republic and Poland) and managed float (the Slovak Republic and Slovenia), which have no announced central parity vis-à-vis the Euro, and the fixed peg not anchored by the Euro (Latvia). The other exchange arrangements—Euro-based currency boards (Estonia and Lithuania) and Euro pegs with horizontal bands (Hungary)—have fixed central parities vis-à-vis the Euro and bands not wider than (15%. From this perspective, they can be considered as valid regime options leading to EMU.

The five countries whose current exchange arrangements are not compatible with EMU entry may switch to a standard ERM-II regime with wide bands, or may go all the way to unilateral euroization. Under the latter option the Euro parity is decided by the accession country unilaterally without agreement from the ECB. This is also the case for currency board arrangements. However, since the ERM-II is a co-operative arrangement between the ECB and the non-euro EU members, the setting of the central parity requires an agreement between the ECB and the accession country in question. From this perspective, currency boards and unilateral euroization can be viewed as non-standard approaches toward EMU.

4.2 EMU Entry via the ERM-II

The standard route toward EMU is via ERM-II. The CEE accession countries willing to take this route need to decide on the timing of their entry into ERM, which influences the timing of EMU assessment and that of the adoption of the euro. There are also technical issues that need to be solved, including the determination of the central parity and the band width. Before we discuss these issues faced by the CEE accession countries, we first take a look at the experience of the first-wave EMU members before they joined EMU on January 1, 1999. These experiences are relevant for the CEE accession countries because the first-wave EMU members all entered into EMU via the standard route of ERM-II.

Past Experiences

On May 1, 1998 the European Council announced that eleven EU member states fulfilled the Maastricht Criteria on nominal convergence and would participate in the third stage of Economic and Monetary Union on January 1, 1999. The relevant assessment for EMU entry was made in March 1998. Data for 1997 was used for the evaluation of fiscal performance. The assessment of inflation and interest rates was based on data for the year ending January 1998. The two-year window for ERM participation referred to the period from March 1996 to February 1998.

Besides a relatively loose interpretation of the criteria pertaining to fiscal deficits and public debts, the assessment also treated the requirement of a two-year ERM participation in a flexible way. Finland joined, and Italy rejoined, the ERM in October and November 1996 respectively. When the assessment was conducted in March 1998, they had not been in the ERM for two years, but they did meet this requirement by the time EMU started. The European Council nevertheless concluded that both countries had displayed sufficient exchange rate stability within the past two years and fulfilled the relevant criterion, since both the Finnish markka and the Italian lira had appreciated vis-à-vis other ERM currencies between March 1996 and their ERM entry, and were stable within ERM afterwards.[31] This leads to two possible interpretations. Either one may conclude that what is important is to avoid devaluations within the two-year period; whether a country is formally participating in the ERM for the whole period is of less importance. Or one may conclude that formal ERM-II does matter, but the relevant period is between the entry to the ERM-II and the entry into the euro. That is, the decision to permit that entry can be made substantially before two years have elapsed (8 months, or 32% of the required period in the example). This is significant, since the announcement of a date for EMU entry and a terminal exchange rate will stabilize the exchange rate and make it easier to meet the criterion.

Setting the terminal conversion rates of the national currencies into the euro was difficult, for the first EMU member states, because it had to meet two conditions set by the Maastricht Treaty. First, the conversion rates had to be equal to the market exchange rates of the national currencies on the last day of currency trading before the start of EMU. Second, the conversion was not allowed to change the value of the national currencies vis-a-vis the ECU, a currency basket of the 15 national currencies including those of the countries with a derogation from EMU. Since the exchange rates between the “ins” and the “outs” were not fixed, the exchange rates of the “ins” into the ECU or the Euro could not be fixed before 1 January 1999, either. Leaving the announcement of the conversion rates until the very last moment could have heightened speculative pressures and create exchange rate volatility in the run-up to EMU that might have derailed the entire process. This risk was particularly large in view of the fact that setting the conversion rates was necessarily a political decision. The outcome of a bargaining in the European Council over the conversion rates right before the start of EMU would have been highly unpredictable, since each country would have had a large incentive to take the others hostage and demand a favorable conversion rate, i.e. one giving its economy a competitive advantage.[32]

In the end, the countries that first joined EMU adopted a solution that promised to minimize uncertainty and exchange rate volatility in the critical phase right before the start of EMU. Specifically, the finance ministers announced already in 1997 that the cross rates among the participating currencies should be those implied by the central parities in the ERM. This provided clear anchors to guide market expectations and closed the room for opportunistic strategic manipulation of the conversion rates (Begg et al., 1997). This decision was confirmed by the European Council in May 1998, which decided that the EMU member currencies would be converted into each other at the fixed central parities prevailing on 1 May 1998. Furthermore, the width of the exchange rate band was left unchanged allowing for maximum fluctuations of 15 percent around the target rate in both directions. Begg et al. (1997) argued that this would minimize the risk of speculative attacks on the central parities, as speculators could not engage in one-sided bets. Nevertheless, the central banks made clear to the markets that they were ready to intervene in unlimited amounts to achieve the central parities in the last trading period before the start of EMU. As it were, the actual transition to EMU proved to be smooth and stable, giving credit to the strategy adopted.

In an interesting theoretical study, De Grauwe et al. (1999) confirm the stabilizing properties of a strategy for conversion into the euro that combines a pre-announced terminal rate with wide exchange rate bands. The key insight is provided by the forward-looking nature of the exchange rate as an asset price. Consider the basic equation of the monetary approach of exchange rate determination:

[pic]

This says that the current exchange rate (in logs), s, is a weighted average of a fundamental equilibrium rate, x, and the expected exchange rate next period, Etst+1. The latter results from the impact of speculative trading in the foreign currency market, which assures that the expected rate of return on domestic and foreign assets is the same. Let x be constant for simplicity and assume that the exchange rate is announced to be fixed at rate s* after n+1 periods. Then

[pic]

This equation says that, as the date when the exchange rate is fixed approaches, the influence of the terminal rate, s*, on the actual rate becomes increasingly larger. Thus, speculation leads the exchange rate to converge smoothly to the terminal rate. However, this is true only if the announcement of the terminal conversion is credible. As De Grauwe et al. (1999) show, any change in the expected terminal rate results in changes in the current exchange rate, and the resulting fluctuations are stronger than in the absence of the conversion announcement. The authors use numerical simulations to compare the variances of the market exchange rate under three scenarios: (1) free float without any guidance on the final conversion rate or its determination, (2) the final conversion rate being the average of past exchange rates, and (3) the final conversion rate being fixed and known to the market. They find that variances of the market exchange rate is the smallest with a pre-announced fixed conversion rate. Note, finally, that the closer the terminal rate, s*, is to the fundamental equilibrium rate, the less variable will be the actual rate as it approaches the terminal date.

Three points can be drawn from the success of the first-round EMU entry. One is that a country should fix its Euro conversion rate upon the confirmation of its entry into EMU or even before that. For future EMU entrants, the rule that the conversion rate must be a market exchange rate of the last trading day will remain, but its implications are less difficult than in the case of the first members of EMU, since the problem of outside currencies no longer exists. Avoiding uncertainty and the risk of disruptive speculative pressures then suggests that they should announce their terminal Euro parities as early as possible.

Another point is that, as long as the announced conversion rate is credible, the market exchange rate will converge smoothly toward this target with the smallest variance. The readiness of the ECB to intervene in support of currencies in the ERM-II is obviously critical for this. Under the rules of the ERM-II, the ECB is obliged to provide unlimited support at the edges of the band, but it can unilaterally withdraw from intervening, if the monetary stability of the euro is endangered otherwise. Since the target rate for entry into EMU is unlikely to be at the edge of the band, this may create unnecessary uncertainty. Instead, the ECB should make clear its willingness in intervene and support the target rate on the last day of trading before a country enters EMU. Such a signal would deter speculative pressures. It is a credible promise, since it would have no adverse consequences for the ECB’s monetary policy: If the ECB acquires a new member’s currency on the last day of trading at the rate s*, and this is the conversion rate of that currency into the euro, the intervention does not affect the ECB’s monetary base on the first day after the new member has entered EMU. Obviously, such a promise should only be made conditional on the decision that the country in question joins EMU.

Beyond that, the third point is that, after fixing the conversion rate, the market exchange rate should be allowed to move freely during the interim period leading to the EMU entry. Central banks should avoid committing themselves to exchange rate bands smaller than the “normal” ones of +/- 15 percent in the ERM-II, so that speculators can not place one-sided bets against the exchange rate. In other words, both the central bank of the new member country and the ECB should refrain from large-scale interventions prior to the last day of trading before the new member country enters EMU. Such interventions would only signal to the markets that the central banks pursue secret exchange rate targets within the normal bands of the ERM-II and invite destabilizing speculation.

Slow versus Fast Track to EMU

There are four important dates for the CEE accession countries during their quest for EMU membership: the date of EU accession, the date of ERM-II entry, the date of assessment for joining EMU, and the date of EMU membership. The date of EU accession is very likely in the middle of 2004 for these countries. From there on, there is a fast-track and a slow-track scenario for the entry into EMU. Under the fast-track scenario, the new EU members would strive for the earliest possible date to become full members of EMU. Because one precondition for joining EMU is at least two-year long participation in ERM-II without devaluation of the currency against the Euro, in strict terms, such assessment can only be made two years after a country enters ERM-II. Since the ERM-II is a co-operative exchange rate arrangement, the earliest date for participation in ERM-II is that of the EU entry. However, it may take time for the new EU members to discuss with the incumbent members and with the ECB on the appropriate central parities for ERM entry. Allowing a few months for negotiations on that issue, the new EU members may enter ERM-II in the second half of 2004. If they prefer an early entry in May 2004, they need to start the negotiations before they join the EU.

The two-year membership requirement without a devaluation then implies that the assessment regarding the Maastricht criteria can be carried out for these countries in the second half of 2006. However, the past experiences with Finland and Italy suggest that earliest possible date for the assessment of the entry criteria would seem to be in the Spring of 2006, when data on required macroeconomic variables (especially fiscal deficits and public debts) for the year 2005 become available. If these countries pass the examination on convergence performance in accordance with the Maastricht Treaty, they could become formal members of EMU on 1 January 2007.

Under the slow-track scenario, countries could take a more relaxed attitude and aim at full EMU membership only a few years later. This implies that there is no need to join the ERM-II immediately after joining the EU. Countries on the slow track could wait until the conditions for joining the exchange rate system and, later on, for holding the assessment regarding the Maastricht criteria seem right.

The main advantage of the fast track scenario is that the instabilities created by the combination of an intermediate exchange rate system, high capital mobility and the expectation of a “convergence play” can be overcome in relatively short time. The main disadvantage is that the new EU members will have to make extra efforts to meet the Maastricht criteria and prepare for the adoption of the euro. To the extent that this requires solutions for inherited structural fiscal problems and further efforts at reducing inflation, the fast track eliminates the possibility to use monetary policy to ease the macroeconomic consequences of structural reforms. The main advantage of the slow-track scenario is that this possibility can be used to some extent, since the exchange rate can still adjust.

This reasoning suggests that those new EU members, that have already given up the use of monetary policy for domestic policy purposes should find the fast-track option more attractive. For the others, which include Poland, the Czech Republic and Hungary, the slow-track option deserves consideration, the more so, the more they perceive that they can use monetary policy effectively to steer their economies. Commitment to an inflation target would then be an appropriate policy framework. Importantly, these countries should not fall into the trap that led to the collapse of the “New ERM” in the early 1990s.[33] That trap was created by the fact that realignments of the central parities became increasingly impossible for reasons of national political prestige, while at the same time monetary and fiscal policies were not fully consistent with a fixed exchange rate. Market beliefs that monetary union was imminent allowed that situation to prevail for some time. However, when the failure of the Danish voters to ratify the Maastricht Treaty shattered those beliefs, the system collapsed in a series of speculative attacks. The lesson to learn for countries on the slow track is twofold. First, these countries should stay out of the ERM-II as long as the conditions for joining EMU are not yet reachable within a period of two or three years. Second, if they decide to join the ERM-II earlier anyway, realignments should remain technical rather than political actions. This can be ensured by establishing regular (annual) and transparent reviews of the appropriateness of the central parities, conducted jointly with the ECB, and timely parity adjustments, if these reviews suggest such steps.

Choosing a Central Parity

Our discussion above and the experience of the first EMU members suggest that market forces will cause the nominal exchange rate of a prospective EMU member’s currency to converge smoothly toward the pre-announced terminal conversion rate. The actual value of the terminal rate is not of great importance from that perspective. However, once the terminal rate has been reached, and given that output prices and wages are sticky, the choice of the nominal exchange rate determines the country’s real exchange rate at least for a while. If the terminal rate for the euro is set too low (in terms of national currency units per euro), the country will enter EMU with an overvalued real exchange rate and experience a period of price level adjustment subsequently, in which the domestic price level falls relative to the price level in the rest of EMU. If the terminal rate is chosen too high, the country will enter EMU with an undervalued real exchange rate and will subsequently experience higher rates of price increases than the rest of EMU. Both deviations from the equilibrium real exchange rate can have significant real costs. In the first case, the industries lose competitiveness and go through an adjustment recession with low growth and rising unemployment. In the second case, over-employment will be the result. As the experiences of Ireland and the Netherlands, among others, suggest, the adjustment may also result in property price bubbles which ultimately can have adverse effects on the financial system.

In addition, the choice of the central parity can also affect a country’s ability to meet the nominal convergence criteria. A country choosing an undervalued central parity when entering the ERM-II may experience high consumer price inflation rates that violate the inflation criterion for EMU participation, unless the actual exchange rate is allowed to adjust within the normal ERM-II band. But if the expectation of EMU entry drives the nominal rate towards the central parity, this adjustment does not take place. Under such circumstances, the exchange rate criterion and the inflation criterion could be conflicting goals.[34] If the market perceives the entry rate as appropriate, the transition into the ERM-II will be easy. Such rates are also viewed as credible, which in turn facilitate smooth convergence of market rates toward the announced parities.

In view of this, the choice of the initial central parity in the ERM-II is of particular importance for the countries on the fast track to EMU, as this parity is likely to be their terminal rate as well. However, equilibrium (real) exchange rates are difficult to determine. The difficulties are further augmented by the fact that the real exchange rate of the new member countries cannot be expected to be a constant. Thus, choosing the central parity is faced with considerable uncertainty.

In principle, there are two ways to resolve this uncertainty. One is to fix the nominal exchange rate for some time and let goods prices adjust to find an equilibrium real exchange rate. The other is to use monetary policy to control domestic inflation and let the nominal exchange rate adjust to find an equilibrium. Which of these is the appropriate solution depends on a country’s monetary policy regime before entering the EU. Those countries that have followed a fixed-exchange rate regime for some time already should use the first option and enter EMU on the fast track, using their past parity to the euro as a guide to pick the central parity in the ERM-II. Should these countries experience rates of price level increase in excess of the inflation criterion, the assessment of nominal convergence should note that this cannot be due to too lax monetary policy, i.e., it reflects real exchange rate adjustments. Since there are then no good reasons to deny these countries entry into EMU, the assessment should make the point transparent and permit their entry.

In contrast, countries that have adopted flexible exchange rates in the recent past should focus on inflation targeting and postpone entry to the ERM-II until domestic inflation is in line with EMU inflation. To the extent that they experience trend appreciations of their currencies during this period, this trend could be accounted for in the choice of a central parity, when these countries eventually decide to initiate the process of entering EMU.

5 Non-Standard Routes toward EMU

5.1 EMU Entry via Currency Board Arrangements

While standard ERM-II regimes have wide fluctuation bands (up to (15%) around the agreed central parities, narrower bands can also be declared as a unilateral commitment of the non-euro area national central bank. As an extreme case, the Euro-based currency board arrangements (CBAs) have the band width set to zero. Another feature that makes CBAs different from standard ERM-II regimes is that their central parities were determined unilaterally by the relevant accession countries without agreement of the ECB, since the CBAs were established well before these accession countries join the EU.

Nevertheless, the existing central parities and the zero band width of the CBAs are likely to be accepted by the ECB as relevant parameters for the ERM-II entry. The regime itself is perfectly compatible with the idea of monetary union with the EU, since all the CBAs now use the Euro as the back-up currency. The central parities have been stable since the launch of the arrangement, reflecting their appropriateness as the exchange rates between accession countries’ national currencies and the Euro, as well as the ability of the monetary authorities of these countries to stabilize the exchange rates without severe pressures. Therefore, the existing Euro parities should be used as the conversion rates upon entry into EMU. Any deviations from the existing parities would only result in unnecessary redistribution of wealth and damages to the established credibility of the regimes. Similarly, introducing broader bands now that the credibility of the CBAs is proven would only create unnecessary uncertainty. Since it is a unilateral commitment, there is no need for the ECB to intervene on the foreign exchange market to support the relevant currencies, hence no risk to price stability in the Euro zone from these arrangements.

With a CBA, the money supply is determined solely by reserve changes, and there is no independent monetary policy instrument to control inflation. Fiscal policy may take this responsibility, but its efficacy is rather limited. Given that the currencies of the accession countries tend to appreciate in real terms against the Euro, the accession countries with CBAs will likely exhibit higher inflation rates than in the Euro area. Thus, although countries with Euro-based CBAs have already effectively adopted the ECB’s monetary policy, there is a risk that they violate the Maastricht criterion on inflation because of the B-S effect. For these countries, any deviation from euro-area inflation should be properly interpreted as relative price adjustment rather than an indicator of weak monetary discipline. Since they have proven their willingness and ability to maintain an absolutely fixed exchange rate with the euro, the inflation criterion should be used in a flexible way when their readiness for EMU is tested.[35]

5.2 EMU Entry via Unilateral Euroization

Unilateral euroization has been proposed by some researchers as a safer step stone toward EMU than the semi-fixed, crises-prone ERM-II regime.[36] With unilateral euroization, a country completely gives up its national monetary policy in favour of that of the ECB. However, it is unable to influence the monetary policy of the ECB before it formally joins EMU, which requires the fulfilment of all Maastricht criteria.

The main argument in favor of unilateral euroization is that it eliminates all room for speculative attacks on a currency in the run up to joining EMU. The combination of large capital inflows and high capital mobility means that exchange rate targets are extremely vulnerable to shifts in market expectations. If market expectations depend on economic fundamentals only, this threat can be regarded as a useful disciplinary device. It forces governments and central banks to maintain safe and sound policies; as long as this is the case, they need not be afraid of being punished by the markets. But if market expectations also depend on sentiments and beliefs independent of fundamentals, the possibility of self-fulfilling and contagious speculative attacks arises. Fixed exchange rates might come under attack because there are problems in neighbouring countries, even if domestic policies are fully consistent with the exchange rate target. From this perspective, sound domestic policies are a necessary but not a sufficient condition for avoiding speculative attacks. Even if governments and central banks adopt all the right measures, a country could be thrown off the path to EMU when a currency crisis spills over from abroad. Empirical and theoretical research to date does not allow us to rule out such effects. How important the threat is, therefore, depends on the degree of risk aversion of the policy makers.

The impact of unilateral euroization on accession countries’ convergence performance is not clear-cut. For real convergence, the impact is likely positive, since a common currency will generally promote trade growth and encourage capital inflows, which are growth-enhancing.[37] For nominal convergence, in contrast, the nature of the impact differs across various aspects. With unilateral euroization, the criterion of exchange rate stability is automatically fulfilled. Unilateral euroization has positive influence on the fulfilment of the interest rate criterion, since the currency risk premium is eliminated. The country risk premium can also be reduced, if euroization eliminates currency mismatches in the foreign liability and, as a result, default risks.[38] However, risk premia may also rise if the euroization is unsustainable due to policy inconsistency or lack of readiness for monetary union.[39] In that case the fulfilment of the interest rate criterion is made more difficult. The impact on fiscal policy is ambiguous, too. On the one hand, fiscal discipline will be strengthened, leading to low deficits and small debt stocks, as governments realize that they can no longer rely on monetary financing any more. On the other hand, a fall of the interest rate reduce the cost of borrowing, which allows governments to pursue lax fiscal policies. As far as inflation is concerned, unilateral euroization deprives the countries of the possibility of nominal appreciation in case of strong real appreciation, so inflation rate at home may well exceed that prevailing in the EU. No matter how the convergence performance is affected, the point is that it is still possible to check the fulfilment of the Maastricht Criteria even if a country has euroized. There is no reason to worry that unilateral euroization would be a cheap way of sneaking into the Euro zone.

A major problem with unilateral euroization is the opposition of the EU to this strategy for EMU entry. Obviously the EU loses its influence on the selection of the rates at which the national currencies are converted into the Euro. However, since currency board arrangements are accepted by the EU despite of their unilaterally selected central Euro parities, there is no reason to object unilateral euroization on this ground. More important is the concern that unilateral euroization might be used as a substitute for serious reforms and to circumvent convergence efforts. This concern is nevertheless not well grounded, as our previous discussions show that the convergence performance of the euroized countries can still be examined based on the Maastricht Criteria. And only by fulfilling all the convergence criteria can a country be granted the membership in EMU.

However, just like with a currency board arrangement, inflation rate convergence is difficult to achieve in a euroized accession country in the presence of trend equilibrium real appreciation. As a result, accession countries adopting unilateral euroization are advised to persuade the EU into adjusting the inflation criterion to make it easier for the accession countries to fulfil.[40] It is not clear whether the EU will soften its position on this issue, which currently insists on applying an equal treatment on the new entrants as on the previous ones. If the convergence criteria remain unchanged, the route toward EMU via unilateral euroization might be a difficult one for the accession countries.

6 Conclusions

In this paper we analyse exchange rate policies in the CEE accession countries during the run-up to EMU. We argue that these countries are confronted with challenges associated with the EU and EMU accession. On the one hand, capital account liberalization, coupled with brighter economic future, generates substantially larger capital flows, leading to high volatility in the foreign exchange market, unless the exchange rate is fixed under a CBA or unilateral euroization. On the other hand, the accession to EMU requires the fulfilment of various convergence criteria, including exchange rate stability and low inflation rates. Since the currencies of the accession countries tend to appreciate in real terms due to relative productivity growth, there is an apparent conflict between the Maastricht Criteria and the economic reality of the accession countries.

Given real appreciation in place, different exchange rate regimes during the run-up to EMU have different implications for nominal convergence. Currency board arrangements ensure exchange rate stability, but price stability is not guaranteed, though fiscal policy can be of some use in attenuating real appreciation. In case of large real appreciation, the inflation target is likely to be missed. A more extreme regime choice is unilateral euroization, which completely eliminates exchange rate risk and reduces interest rates by lowering risk premia. However, like under CBAs, inflation tends to be higher under euroization than under a standard ERM-II regime. The suggested solution is to soften the inflation criterion to make it compatible with the situation of the CEE countries.

The standard choice is to enter the ERM-II with a fixed central parity and a wide fluctuation band. A wide band is necessary to accommodate exchange rate volatility associated with enhanced capital mobility. The Euro central parties during the ERM-II should be, in normal cases, the final conversion rates between national currencies and the Euro. These Euro parities, as well as their importance as the predecessor of the final Euro conversion rates, should be made clear to the markets as early as possible, as this can reduce much confusion in the prediction of the conversion rates and much fluctuation in the market rates as they converge toward the announced targets. However, if the pressure of real appreciation is strong, the central parities can be revalued to ease the fulfilment of the relevant nominal convergence criteria.

References

Amato, Jeffrey D., and Stefan Gerlach, 2001, “Inflation Targeting in Emerging Market and Transition Economies: Lessons After a Decade”, CEPR Discussion Paper No. 3074.

Backé, Peter, Jarko Fidrmuc, Thomas Reininger, and Franz Schadarx, 2002, “Price Dynamics in Central and Eastern European EU Accession Countries”, Oesterreichische Nationalbank Working Paper 61.

Bank of Greece, 2000, Monetary Policy. Interim Report 2000.

Begg, David, Francesco Giavazzi, Jürgen von Hagen, and Charles Wyplosz, 1997, “EMU: Getting the End-game Right”, Monitoring European Integration 7, CEPR.

Begg, David, Barry Eichengreen, László Halpern, Jürgen von Hagen, and Charles Wyplosz, 2001, Sustainable Regimes of Capital Movements in Accession Countries, CEPR.

Berg, Andrew, and Eduardo Borensztein, 2000, “The Pros and Cons of Full Dollarization”, IMF Working Paper 00/05.

Buch, Claudia M., and Lusine Lusinyan, 2002, “Short-term Capital, Economic Transformation, and EU Accession”, Deutsche Bundesbank Discussion Paper 02/02.

Buiter Willem H., and Clemens Grafe, 2001, “Central Banking and the Choice of Currency Regime in Accession Countries”, SUERF Studies No. 11.

Buiter Willem H., and Clemens Grafe, 2002, “Anchor, Float or Abandon Ship: Exchange Rate Regimes for Accession Countries”, CEPR Discussion Paper No. 3184.

Coricelli, Fabrizio, and Bostjan Jazbec, 2001, “Real Exchange Rate Dynamics in Transition Economies”, CEPR Discussion Paper No. 2869.

De Broeck, Mark, and Torsten Sløk, 2001, “Interpreting Real Exchange Rate Movements in Transition Countries”, BOFIT Discussion Paper 2001 No. 7.

De Grauwe, Paul, Hans Dewachter, and Dirk Veestraeten, 1999, “Price Dynamics under Stochastic Process Switching: Some Extensions and an Application to EMU”, Journal of International Money and Finance 18 (1999), pp. 195—224.

Deutsche Bank Research, 2002, EU Enlargement Monitor.

ECOFIN, 1998, “Recommendation on Member States Participating in EMU”, Brussels.

Eichengreen, Barry, 1994, International Monetary Arrangements for the 21st Century, Washington DC: Brookings Institution.

European Bank for Reconstruction and Development, Transition Report (various issues).

European Union, 2002, Presidency Conclusions. Copenhagen European Council, available from

European Commission, 2002, Strategy Paper, available from

European Commission, 2002, Regular Report, available from

Fischer, Christoph, 2002, “Real Currency Appreciation in Accession Countries: Balassa-Samuelson and Investment Demand”, Deutsche Bundesbank Discussion Paper 19/02.

Fischer, Stanley, 2001, “Exchange Rate Regimes: Is the Bipolar View Correct?” mimeo, IMF.

Fratianni, Michele, and Jürgen von Hagen, 1992, The European Monetary System and European Monetary Union, Boulder and Oxford: Westview Press.

Ghosh, Atish R., Anne-Marie Gulde, and Holger C. Wolf, 2000, “Currency Boards: More Than a Quick Fix?” Economic Policy October 2000, pp. 271—335.

Goldstein, Morris, 2002, Managed Floating Plus, Washington, DC: Institute for International Economies.

Halpern, László, and Charles Wyplosz, 1997, “Equilibrium Exchange Rates in Transition Economies”, IMF Staff Papers, Vol. 44, No. 4, pp. 430—461.

Halpern, László, and Charles Wyplosz, 2001, “Economic Transformation and Real Exchange Rates in the 2000s: The Balassa-Samuelson Connection”, mimeo.

Ho, Tai-kuang, 2003, Four Essays in the Economics of Banking and Currency Crises, PhD dissertation, ZEI, University of Bonn.

International Monetary Fund, 2001, “Monetary and Exchange Rate Regimes in the Central European Economies on the Road to EU Accession and Monetary Union”, mimeo.

International Monetary Fund, Annual Report on Exchange Arrangements and Exchange Restrictions (various issues).

Kaminsky, Graciela, and Carmen Reinhart, 1999, “The Twin Crises: The Causes of Banking and Balance-of-Payments Problems”, American Economic Review, 89 (3), pp. 473—500.

Kopits, George, and István Székely, 2002, “Fiscal Policy Challenges of EU Accession for the Baltics and Central Europe”, mimeo, IMF.

Kovács, M. András, and András Simon, 1998, “Components of the Real Exchange Rate in Hungary”, National Bank of Hungary Working Paper 1998/3.

Lommatzsch, Kirsten, and Silke Tober, 2002, “Monetary Policy Aspects of the Enlargement of the Euro Area”, Deutsche Bank Research, Research Notes No. 4.

Mussa, Michael, Paul Masson, Alexander Swoboda, Esteban Jadresic, Paolo Mauro, and Andrew Berg, 2000, “Exchange Rate Regimes in an Increasingly Integrated World Economy” IMF Occasional Paper 193.

Obstfeld, Maurice, and Kenneth Rogoff, 1996, Foundations of International Macroeconomics, Cambridge, MA: MIT Press.

Orlowski, Lucjan T., 2000, “A Dynamic Approach to Inflation Targeting in Transition Economies”, ZEI Working Paper B00-11, University of Bonn.

Pelkmans, Jacques, Daniel Gros, and Jorge N. Ferrer, 2000, “Long-run Economic Aspects of the European Union’s Eastern Enlargement”, WRR Working Documents W109.

Rostowski, Jacek, 2002, “Why Unilateral Euroisation Makes Sense for (Some) Applicant Countries”, mimeo.

Rother, Philipp, 2000, “The Impact of Productivity Differentials on Inflation and the Real Exchange Rate: An Estimation of the Balassa-Samuelson Effect in Slovenia”, IMF, mimeo.

Szapáry, György, 2000, “Maastricht and the Choice of Exchange Rate Regime in Transition Countries During the Run-up to EMU”, National Bank of Hungary Working Paper 2000/7.

Von Hagen, Jürgen, and Jizhong Zhou, 2002, “The Choice of Exchange Rate Regimes: An Empirical Analysis for Transition Economies”, ZEI Working Paper B02-03, University of Bonn.

-----------------------

* This paper is prepared for the conference on Monetary Strategies for Accession Countries, February 27—28, 2003, Budapest, organized jointly by the National Bank of Hungary (NBH), the Institute for World Economics of the Hungarian Academy of Sciences (IWE), and the Center for European Integration Studies (ZEI), University of Bonn.

** Correspondence: ZEI, Walter Flex Strasse 3, D-53113 Bonn, Germany. Tel: +49+228+734928.

Email: zhou@united.econ.uni-bonn.de

[1] European Union (2002).

[2] Recent currency crises in EMS (1992), Mexico (1994), East Asia (1997), Russia (1998), Brazil (1999), and Turkey (2000) all involved some type of adjustable exchange rate peg in a financially open economy. See Eichengreen (1994), Fischer (2001), or Mussa et al. (2000) for more discussions on the fragility of semi-fixed exchange rates under high capital mobility.

[3] Halpern and Wyplosz (1997, 2001).

[4] See von Hagen and Zhou (2002) for an empirical analysis of the exchange rate regime choices in transition countries in the 1990s.

[5] For example, in 2001 the average inflation rate in the eight CEE accession countries is 5.7%, while the inflation rate in the Euro area is 2.6%.

[6] In 2001 the GDP values of Estonia and Lithuania, measured in purchasing power standard, are 13.4 billion Euro and 30.3 billion Euro, respectively, ranking number 8 and 6 in the eight CEE countries and much below the group average of 96 billion Euro.

[7] See, among others, Ghosh et al. (2000). Another regularity is that CBAs tend to adopted by countries with very poor track record in macroeconomic policy management. A case in point from the CEE countries is Bulgaria, which adopted the CBA in 1997 in the aftermath of a severe currency crisis and a hyperinflationary episode.

[8] The Lithuanian currency board originally used US dollar as the back-up currency. To make the exchange rate regime more compatible with the requirement of the EU accession, it has been replaced by the Euro as the reserve currency since February 2, 2002.

[9] Amato and Gerlach (2001) point out that central bank independence, especially instrument independence, is an essential precondition for the success of an inflation targeting framework. Goal independence, in contrast, is less important, as the UK experience suggests.

[10] The EBRD indices are valued on a 1,2,3,4 scale, with “x-” or “x+” indicating a value slightly below or above x, and with “4+” denoting full convergence with the standards and norms of advanced industrial countries. To derive the index reported in the text, a “+/-” perturbation is interpreted as “0.3 more or less”, i.e., “3-” is valued as 2.7 and “3+” as 3.3. The EU reference value is set at 4.3.

[11] Orlowski (2000) argues that the CEE accession countries should gradually replace inflation targeting by exchange-rate targeting when they are preparing for the final entry into EMU.

[12] IMF (2001).

[13] The so-called “Copenhagen Criteria” for the accession into the EU require that accession countries ensure (1) stability of institutions guaranteeing democracy, the rule of law, human rights and the respect and protection of minorities; (2) the existence of a functioning market economy as well as the capacity to cope with competitive pressure and market forces within the Union; and (3) ability to take on the obligations of membership, including adherence to the aims of political, economic and monetary union.

[14] One obvious disadvantage of focusing on per capita data is the negligence of income distribution.

[15] For example, the average PPP-adjusted per capita GNP of Greece, Ireland, Portugal, and Spain is 55% of the EU average in 1986, and 69% of the EU average in 1997. See Buiter and Grafe (2001).

[16] Simple average based on 2001 data.

[17] These numerical targets should be interpreted with flexibility. Larger deficits or public debts may be allowed as long as they are falling substantially and approaching the reference values at a satisfactory speed.

[18] See Buch and Lusinyan (2002).

[19] Since 1997 the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions identifies eleven categories of capital transactions that may be subject to controls. They include controls on (1) capital market securities, (2) money market instruments, (3) collective investment securities, (4) Derivatives and other investments, (5) commercial credits, (6) financial credits, (7) guarantees, sureties and, financial backup facilities, (8) direct investment, (9) liquidation of direct investment, (10) real estate transactions, and (11) personal capital movements.

[20] See e.g. Kaminski and Reinhart (1999), Ho (2003).

[21] Price liberalization may also contribute to the adjustment of relative prices, but its importance fades away as the scope of price controls becomes increasingly smaller in recent years in the accession countries.

[22] See, among others, Halpern and Wyplosz (2001).

[23] Obstfeld and Rogoff (1996, p. 212) show that the B-S effect can be characterized by the following equation:

À - s, Halpern and Wyplosz (2001).

[24] Obstfeld and Rogoff (1996, p. 212) show that the B-S effect can be characterized by the following equation:

π - π* = (1 - α) [( (aT – aT*) - (aN - aN*)],

where ( denotes relative labour intensity of the non-tradable sector compared to the tradable sector, and variables with asterisk denote foreign counterparts. Assuming zero productivity growth abroad, we have

π - π* = (1 - α)( (aT – aN) + (1 - α)(( - 1)aN,

which leads to the approximation used in the text, if we assume that aN is small, and ( is not far above unity.

[25] This is the simple average of relative productivity growth (aT-aN) in eight CEE accession countrie. For each country aT (aN) is the average productivity growth rate of the tradable (non-tradable) sector during 1994—1999. The tradable sector is proxied by the manufacturing, while the non-tradable sector covers trade, repair, hotel, etc. and financial service and real estate. The data source is UNECE. Note that our estimation of 7.3% is not substantially different from that of Halpern and Wyplosz (2001), whose estimate is 6.7%.

[26] This corresponds to the average trade-to-GDP ratio in the eight accession countries over 1995—1999. The trade value is the average of export and import values.

[27] Note that the 1.5% differentials allowed by the Maastricht Treaty is relative to the three lowest inflation rates in the EU member states, while our estimation refers to an inflation differential relative to the EU average rate, so the chance for the countries to fulfill the inflation criterion is even less.

[28] See Backé et al. (2002) and Lommatzsch and Tober (2002).

[29] Rother (2000) reports that the relative productivity growth in the EU can offset the inflation differential by roughly 0.8%.

[30] See Fischer (2001) and Eichengreen (1994).

[31] See Begg et al. (2001).

[32] ECOFIN (1998).

[33] Note that the economic cost of such implied devaluation in terms of inflation could be expected to be small, since the ECB was to take over monetary policy and fight inflation immediately afterwards.

[34] The “New ERM,” a term used in the monetary policy debates of the late 1980s and early 1990s, was built on the assumption that, after 1987, countries would not use the option of realignments any more and that the ERM, therefore, was already a de-facto monetary union with some remaining exchange rate flexibility. See Fratianni and von Hagen (1992).

[35] This discussion implies that the strategy adopted by Greece, which devalued the drachma by 12.3% just before the entry into the ERM in March 1998 bears considerable risks. See Bank of Greece (2000).

[36] See Szapáry (2000), Buiter and Grafe (2001), and Begg et al. (2001).

[37] See Begg et al. (2001), Rostowski (2002), Buiter and Grafe (2002).

[38] See Goldstein (2002) and Rostowski (2002).

[39] See Berg and Borensztein (2000).

[40] Begg et al. (2001).

[41] See Rostowski (2002).

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download